feat(Go-Tool)：2020/12/15：添加前缀树工具，实现前缀搜索和字符串搜索功能

util/trie/Trie.go

@@ -0,0 +1,156 @@
+/*
+ *  @Author : huangzj
+ *  @Time : 2020/12/14 16:15
+ *  @Description：前缀树\字典树
+ */
+
+package trie
+
+import "github.com/ahmetb/go-linq"
+
+type Trie struct {
+	Child map[rune]*Trie //存储当前字符（Rune）下面对应的Rune的集合，使用Map可以提高查找效率
+	Word  string         //存储字典树上有的字符串信息,如果Word是空字符串("")，则代表没有在该节点找到对应的字符串信息
+}
+
+const (
+	emptyTag = ""
+)
+
+type WordFunc func(sentence []rune, start int, end int) string
+
+func New() *Trie {
+	return initTrie()
+}
+
+//初始化Trie
+func initTrie() *Trie {
+	return &Trie{
+		Child: make(map[rune]*Trie, 0),
+		Word:  emptyTag,
+	}
+}
+
+//往前缀树中添加对应的字符串(build模式)
+func (trie *Trie) insert(word string) *Trie {
+	traverse := trie
+	for _, v := range []rune(word) {
+		if _, ok := traverse.Child[v]; !ok {
+			newTrie := initTrie()
+			traverse.Child[v] = newTrie
+		}
+		traverse = traverse.Child[v]
+	}
+	traverse.Word = word
+	return trie
+}
+
+//删除对应的字符串，如果查找不到则直接返回(build模式)
+func (trie *Trie) Delete(word string) *Trie {
+	traverse := trie
+	for _, v := range []rune(word) {
+		if _, ok := traverse.Child[v]; !ok {
+			return trie
+		}
+		traverse = traverse.Child[v]
+	}
+	traverse.Word = emptyTag
+	return trie
+}
+
+//查找对应的字符串，找到则返回true，没找到则返回false
+func (trie *Trie) Search(word string) bool {
+	traverse := trie
+	for _, v := range []rune(word) {
+		if _, ok := traverse.Child[v]; !ok {
+			return false
+		}
+		traverse = traverse.Child[v]
+	}
+	if traverse.Word != emptyTag {
+		return true
+	}
+	return false
+}
+
+//遍历前缀树存储的所有字符串
+func (trie *Trie) Traverse() []string {
+	result := make([]string, 0)
+	result = traverse(trie, result)
+	return result
+}
+
+func traverse(trie *Trie, result []string) []string {
+	for _, value := range trie.Child {
+		if value.Word != emptyTag {
+			result = append(result, value.Word)
+		}
+		//进行子的路径遍历
+		if len(value.Child) != 0 {
+			result = traverse(value, result)
+		}
+	}
+
+	return result
+}
+
+//查找字符串中匹配的子字符串，进行自定义处理
+func (trie *Trie) DealFunc(sentence string, wordFunc WordFunc) string {
+	traverse := trie
+	for index, v := range []rune(sentence) {
+		//从句子的字符开始往下查找,找到最后一个对应的字符(如果uuck和uucku都是敏感词汇的话，会把uucku设置为*,而不只是uuck.)
+		if _, ok := traverse.Child[v]; ok {
+			traverse = traverse.Child[v]
+			continue
+		}
+		if traverse.Word != emptyTag {
+			//传入到wordFunc里面的是字符串的rune数组、查询到的词汇在字符串中的开始位置、查询到的词汇在字符串中的结束位置
+			sentence = wordFunc([]rune(sentence), index-len([]rune(traverse.Word)), index)
+		}
+		//trie从头开始
+		traverse = trie
+	}
+
+	return sentence
+}
+
+//搜索前缀匹配的字符串集合(精准匹配)
+func (trie *Trie) SearchTrie(word string) ([]string, bool) {
+	traverse := trie
+	for _, value := range []rune(word) {
+		if _, ok := traverse.Child[value]; !ok {
+			return []string{}, false
+		}
+		traverse = traverse.Child[value]
+	}
+	return traverse.Traverse(), true
+}
+
+//搜索前缀匹配的字符串集合(模糊匹配)
+func (trie *Trie) SearchTrieVague(word string) ([]string, bool) {
+	var tries []*Trie
+	traverse := trie
+	for _, value := range []rune(word) {
+		if _, ok := traverse.Child[value]; !ok {
+			break
+		}
+		traverse = traverse.Child[value]
+		tries = append(tries, traverse)
+	}
+	result := make([]string, 0)
+	for _, item := range tries {
+		result = append(result, item.Traverse()...)
+	}
+
+	if len(result) == 0 {
+		return []string{}, false
+	}
+	realResult := make([]string, 0)
+	linq.From(result).Distinct().Sort(func(i, j interface{}) bool {
+		if i.(string) > j.(string) {
+			return true
+		}
+		return false
+	}).ToSlice(&realResult) //去重
+	return realResult, true
+}